Class SortingStrategy

java.lang.Object
org.apache.lucene.analysis.hunspell.SortingStrategy

public abstract class SortingStrategy extends Object
The strategy defining how a Hunspell dictionary should be loaded, with different tradeoffs. The entries should be sorted in a special way, and this can be done either in-memory (faster, but temporarily allocating more memory) or using disk (slower, but not needing much memory).
See Also:
  • Constructor Details

    • SortingStrategy

      public SortingStrategy()
  • Method Details

    • offline

      public static SortingStrategy offline(Directory tempDir, String tempFileNamePrefix)
      An "offline" strategy that creates temporary files in the given directory and uses them for sorting with OfflineSorter. It's slower than inMemory(), but doesn't need to load the entire dictionary into memory.
    • inMemory

      public static SortingStrategy inMemory()
      The strategy that loads all entries as String objects and sorts them in memory. The entries are then stored in a more compressed way, and the strings are gc-ed, but the loading itself needs O(dictionary_size) memory.